USUM: Update Summary Generation System

TL;DR

The paper introduces USUM, a system for efficiently updating query-specific multi-document summaries by embedding existing summaries into new documents, avoiding re-generation from scratch.

Contribution

It proposes a novel method to update summaries incrementally by embedding previous summaries into new documents, enhancing efficiency and practicality.

Findings

Good performance in quality and efficiency

Effective in scenarios with inaccessible source documents

Outperforms re-generation methods

Abstract

Huge amount of information is present in the World Wide Web and a large amount is being added to it frequently. A query-specific summary of multiple documents is very helpful to the user in this context. Currently, few systems have been proposed for query-specific, extractive multi-document summarization. If a summary is available for a set of documents on a given query and if a new document is added to the corpus, generating an updated summary from the scratch is time consuming and many a times it is not practical/possible. In this paper we propose a solution to this problem. This is especially useful in a scenario where the source documents are not accessible. We cleverly embed the sentences of the current summary into the new document and then perform query-specific summary generation on that document. Our experimental results show that the performance of the proposed approach is good in terms of both quality and efficiency.